July 5, 20178 yr 1 hour ago, SilverNodashi said: Quite some interesting reading... @Coulomb, can't you modify the firmware to fix the problem? Possibly, given enough time (I assume you are referring to the fan locked rotor problem here). There's just not enough incentive to justify the time required.
July 5, 20178 yr Author 28 minutes ago, Chris Hobson said: go figure! yours sounds like a firmware problem, the axpert doesn't seem to pick up that there is enough energy in the batteries to switch to them - instead it happily draws power from the generator whilst wasting your fuel, especially if the gennie has a automatic switch-on from the inverter. there was a question re ROI - well, under these circumstances getting there will take much longer; yours with fuel wastage and mine with parts starting to fail [just after warranty expiration as usual] whilst being on a different level than my situation, it is, as you say strange behaviour - wish we could get someone from mercer/axpert on the forum to answer these 'anomalies' - or maybe there already is someone?
July 5, 20178 yr 1 hour ago, Coulomb said: Possibly, given enough time (I assume you are referring to the fan locked rotor problem here). There's just not enough incentive to justify the time required. Yes. That's what I was referring to.
July 5, 20178 yr 52 minutes ago, gabriel said: yours sounds like a firmware problem, the axpert doesn't seem to pick up that there is enough energy in the batteries to switch to them - instead it happily draws power from the generator whilst wasting your fuel, especially if the gennie has a automatic switch-on from the inverter. there was a question re ROI - well, under these circumstances getting there will take much longer; yours with fuel wastage and mine with parts starting to fail [just after warranty expiration as usual] whilst being on a different level than my situation, it is, as you say strange behaviour - wish we could get someone from mercer/axpert on the forum to answer these 'anomalies' - or maybe there already is someone? No it is set to run off the gennie. The gennie is manually switched on and runs for other purposes and it makes more sense to use gennie power for 2-3 hours than be powering the house off batteries. I don't think it is anomalous behaviour - just I would expect the temperature to drop and it does not. I just need to understand why.
July 5, 20178 yr Author sorry, didn't know you were actually referring to the temp issue - yes, indeed, if the house is on the generator... but does the current pass through the inverter in some way? for the increased temperature would indicate that the electricity from the gennie is actually encountering resistance from the inverter - or would it be that it has something to do with converting it to 'pure sine wave'... just throwing around some ideas as, as stated before, electronics is not my forte...
July 5, 20178 yr 9 minutes ago, gabriel said: actually encountering resistance from the inverter - or would it be that it has something to do with converting it to 'pure sine wave'... Naaah, the inverter is probably float-charging the batteries from AC (right?) so the issue would be that it seems to get disproportionately hot from doing this task (compared to when it runs as an inverter). The 5kva model is also a little different to the smaller units, I think the 3kva (and probably the 24V one) has a separate charger, whereas the 5kva runs the boost stage in reverse to do charging. All in all... it's probably normal, or something.
July 5, 20178 yr Author 2 minutes ago, plonkster said: or something i'll settle for that and have a cuppa...
July 5, 20178 yr 50 minutes ago, plonkster said: Naaah, the inverter is probably float-charging the batteries from AC (right?) so the issue would be that it seems to get disproportionately hot from doing this task (compared to when it runs as an inverter). The 5kva model is also a little different to the smaller units, I think the 3kva (and probably the 24V one) has a separate charger, whereas the 5kva runs the boost stage in reverse to do charging. All in all... it's probably normal, or something. I'd go with "or something." But ^^^ this is my guess as well. The high frequency inverter is probably the hotest component, and as had been established the inverter use the highest temp reading of the 3 thermistors, so I'm guessing the one closest to the high frequency inverter is "the culprit". You could confirm this with a thermal camera or thermal laser gun - borrow / buy / steel one somewhere if you have to.
July 5, 20178 yr Author despite cool advice, things are heating up on this thread ;-) 25 minutes ago, SilverNodashi said: thermal camera or thermal laser gun were really getting into the star wars gear now, just kidding - it would be very interesting to see the verdicts on this one in the meantime back in peacefortress i surrounded the top of the axpert with three old pc fans [powered by old 12v powersupply, 1.5A] with doublesided tape, sucking air out of the unit - resulting in a near immediate 9C drop!!!
July 5, 20178 yr 52 minutes ago, gabriel said: despite cool advice, things are heating up on this thread ;-) were really getting into the star wars gear now, just kidding - it would be very interesting to see the verdicts on this one in the meantime back in peacefortress i surrounded the top of the axpert with three old pc fans [powered by old 12v powersupply, 1.5A] with doublesided tape, sucking air out of the unit - resulting in a near immediate 9C drop!!! Nice! So now you don't void the warrantee, though the double sided tape residue might be questionable I doubt if the top fan would do much though. Next step would be to put an Arduino on there with it's own themistor and only let the fans come in when the case is too hot - but then you need to find the hottest point, and play around with the code a bit. this could possibly save a couple watts during the night. Unless you run those fans off a 7A or 18A alarm battery with it's own 10W PV panel or something?
July 5, 20178 yr Author @SilverNodashi 53 minutes ago, SilverNodashi said: Next step would be ... its about there where you lost me i realize the top fan is of hardly any use, i'll remove it - the small power supply [ex router or something, see snap] will be able to relax a bit. i don't worry about the tape residue... zer ar vayz und meanz [pardon my germlish]
July 5, 20178 yr 19 minutes ago, gabriel said: @SilverNodashi ... its about there where you lost me Is that grammar that bad? Just kidding. I guess the two extractor fans could run constantly bur their lifespan would probably be shorten. Though they only cost a couple bucks to replace.
July 5, 20178 yr Author nothing wrong with your grammar, its just that i somehow made it through 'school' without attending those classes - there are other classes i also missed, to my detriment i confess :-) indeed those fans are cheap, between 50 and 150 bucks new and the ones in my pc have been at it constantly for about 5 years so i'll let them go 24/7. the lowest temp my inverter ever hit [apart for a 'glitch' of 39] has been 43, like at 03:00, now it is comfortably cruising at 46; but let us not get exited, the proof of the pudding will only be visible when the grid kicks in to charge the batteries later today when customary peaks of 70 were common. i'll post feedback at that time. cya
July 5, 20178 yr 4 hours ago, SilverNodashi said: in the meantime back in peacefortress i surrounded the top of the axpert with three old pc fans [powered by old 12v powersupply, 1.5A] with doublesided tape, sucking air out of the unit - resulting in a near immediate 9C drop!!! Can I suggest that you blow air in from the top! This may sound counter-intuitive but the air flow is out the bottom of the inverter without the fans reversed and now we have air out both top and bottom and no real airflow in. Yes it will get in somewhere but you may just end up with a dead spots due to turbulent airflow.
July 5, 20178 yr Author thanks @Chris Hobson, i'll try that - it makes sense. in the meantime though feedback on the present situation. max temp during load came down from 69/70 to 65; whereas in the past it dropped to about 53/55 it is at present 41 [i.e. when load dropped]. conclusion: there is meaningful improvement! now for top fan reversal...
July 5, 20178 yr Author not knowing the inside 'aerodynamics' of the axpert i did the following, see jpg prior to experiment both top left and top right fans blowed outward - result 41C top right in top left out - 54C [skipped opposite] top left in top right in as @Chris Hobson suggested - 47/46C took top left off, top right out - 46C [skipped opposite] back to both out - 41C, now, 60 odd min later it is at 39C if the thermistors are placed on or very near the heat bearing elements moderate airflow should not have too much influence on them. we should take note that the heat sink is at the top, designers must have anticipated correctly that heat would rise and should be dissipated from there, something the top fans are in fact now doing [but artificially and in so doing 'flat spots' might be created, i admit]. i have repeatedly touched the heat sink not feeling any warmth... maybe it is very effective? the other thing is, will the reversal of the bottom fan not also create 'flat spots'?
July 6, 20178 yr 13 hours ago, SilverNodashi said: Next step would be to put an Arduino on there with it's own themistor and only let the fans come in when the case is too hot - but then you need to find the hottest point, and play around with the code a bit. this could possibly save a couple watts during the night. Unless you run those fans off a 7A or 18A alarm battery with it's own 10W PV panel or something? I would not even go that fancy . I used to build these for my "gaming" PC many moons ago it works like a charm and is as low tech as what you can get for fan temperature control.. http://myclassbook.org/temperature-controlled-dc-fan-using-thermistor/
July 6, 20178 yr Morning Gabriel Having conducted your tests I would say keep them in your original orientation. Just as a final test see what happens under load when the Axpert's own fans are running too. 6 hours ago, gabriel said: we should take note that the heat sink is at the top, designers must have anticipated correctly that heat would rise and should be dissipated from there, something the top fans are in fact now doing [but artificially and in so doing 'flat spots' might be created, i admit]. i have repeatedly touched the heat sink not feeling any warmth... maybe it is very effective? The heatsink at the top serves only the SCC. It should be warm to hot during the day and cool at night. A week ago I would not have been able to tell you that but I now have had the advantage of tearing @Virgil's Axpert apart.
July 6, 20178 yr Author 15 hours ago, SilverNodashi said: though the double sided tape residue might be questionable plus on the inside of the box there will be dust accumulation in the form of the fans.... the 'locard principle' WILL prevail 14 minutes ago, Chris Hobson said: The heatsink at the top serves only the SCC. It should be warm to hot during the day and cool at night. well yes, i haven't had time to hold on to it for extended periods which i presume a min/max thermometer would be able to have a blessed day all you people out there! in Christ gabriel
July 6, 20178 yr 9 minutes ago, gabriel said: well yes, i haven't had time to hold on to it for extended periods which i presume a min/max thermometer would be able to When I remember I just run my hand over it. Summer time it nearly gets too hot to touch.
September 8, 20178 yr Hi @gabriel My Axperts are running at around 70'C during the day on solar (very seldom require grid at this point), what was your final fan orientation that gave best results? I would like to do something similar.
September 8, 20178 yr On 7/6/2017 at 5:48 AM, PaulF007 said: http://myclassbook.org/temperature-controlled-dc-fan-using-thermistor/ Gah, there's no hysteresis on that circuit... :-) For those who don't know: You should place a resistor from the output of the op-amp (pin 6) to the positive input (pin 3) to create a bit of positive feedback. What this does is to "exagerate" the voltage difference between the pins at the switch-over point to avoid oscillation on the output. Edit: Hang on, I see the output drives a PNP in emitter-follower mode, so the idea is clearly to control the fan speed. In that case it should have a resistor going to the negative pin to control the gain :-)
September 8, 20178 yr Author 40 minutes ago, pilotfish said: what was your final fan orientation that gave best results? I would like to do something similar. two pc fans driven by old [phone charge?] 12v adapter stuck on either side with double sided tape SUCKING AIR OUT OF UNIT . drop in temp at least 10c - no untoward effects [yet] , fans on 24/7, they are cheap enough and last for years
September 8, 20178 yr Thanks @gabriel, I reckon 70'C is a bit over the top in the long haul, so the additional fans will go in over the weekend.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.